Date:         Tue, 18 Dec 2001 10:41:26 +0100
Reply-To:     Asesoría Bioestadística
              <bioestadistica@eresmas.net>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Asesoría Bioestadística
              <bioestadistica@eresmas.net>
Subject:      Re: poisson and negative binomial distribution glm
Content-Type: text/plain; charset=us-ascii

Hi Marcus

Try log-linear models instead of GLM. A good place to find information is
this one:

 http://www2.chass.ncsu.edu/garson/pa765/statnote.htm

HTH

Marta


Marcus Chibucos ha escrito:

> Hello.  I am new to this listserve.  I am performing a GLM with discrete
> (count) data, and would like to specify my model to match the POISSON
> distribution. I'd like to do another analysis (MANOVA) with a NEGATIVE
> BINOMIAL distribution. Is there a way to specify the distributions, or
> must SPSS work on an assumption of NORMAL distribution?  If this could
> be specified in syntax, could someone please tell me how?  Thanks a lot.
> Sincerely,
>
> Marcus Chibucos
>
> Graduate Student
>
> Department of Biological Sciences
>
> Bowling Green State University
>
> Bowling Green, OH 43403